Hi all,
I thank you for your answers.
@all, in my current case, the contributors have a valid CLA, so I don’t have problem about this point
@Emilio, I think I found the same GitHub option in GitLab but it seems forbidden to apply it for MR targeting
a protected branch (like master). I will investigate more about this point.
@Dennis, thank your for your script, we will have a look to it.
On my side, I have qualms about asking the contributor to correct a minor point or just to rebase its code
because we're 2 months behind in our reviews. It takes me more time to ask for a rebase, waiting for it and merge, that doing it myself, that why I was looking for tips. In my opinion, these points were really easy to adress with Gerrit and become very painful
with gitlab.
Cheers,
--
Vincent LORENZO
De : eclipse.org-committers <eclipse.org-committers-bounces@xxxxxxxxxxx>
De la part de James Hughes via eclipse.org-committers
Envoyé : vendredi 24 janvier 2025 13:53
À : Eclipse Committers general discussion <eclipse.org-committers@xxxxxxxxxxx>
Cc : James Hughes <jnhuva@xxxxxxxxx>
Objet : Re: [eclipse.org-committers] -EXT- Do you have tips to update external contributions with Gitlab ?
Hi all,
From my understanding, I'd point out that in Emilio's case there'd need to be a PR and the author would need to have a CLA already signed.
Basically, if you could take the contribution, then I'd say that you are accepting the contribution and then doing something on top of it.
In the past, I've definitely cloned numerous forks of GeoMesa in order to pull PR branches (from folks with a CLA clearly contributing the code) in order to clean up history or make small modifications. Using `git commit --author` can let you reset the attribution
in a commit.
It is only in the last few years that the GitHub UI has added enough options to handle typical situations where it has been useful enough to skip most of those operations. ;) I have no clue how similar workflows work in GitLab.
That said, it is pretty typical for projects to ask folks submitting a PR to fix minor issues and/or to rebase it. Vincent , you are being very nice for wanting to save the contributor this work!
I'm not sure about gitlab, but in GitHub there is an option (selected by default) when opening a PR to allow committers to push changes to your branch. If that's enabled, then committers
can use any of the usual tools to modify the PR branch. But otherwise, you generally can't update someone else's fork. If the author is unresponsive, and the contribution is valuable, then I believe it's generally ok to pull their branch (which must be public
in order to open a PR), push it somewhere you control, and open a new PR, as long as you keep the original author attribution.
WARNING: This message is from an external source. Evaluate the message carefully BEFORE clicking on links or opening attachments.
Dear Committers,
I would like to know how you proceed on your project to fix/update yourself a merge requests received from a fork of your project.
In the Papyrus-Web project, on the git,
https://gitlab.eclipse.org/eclipse/papyrus/org.eclipse.papyrus-web,
we receive MergeRequests from fork looking like this https://gitlab.eclipse.org/USERGIT/org.eclipse.papyrus-web/-/tree/A_BRANCH
.
Please, do you know if it exists an easy way to modify/amend the contents of the MergeRequest ? On my side I didn’t find answer on the web and ChatGPT advice me to update the
fork to update the MR…
I don’t want to have to clone the fork (in addition, it could be private I think) , then to update it to update the MR.
I don’t want to have to refuse the MR for minor things that I can solve by myself. The basic usecase is a trivial rebase, that gitlab is not able to do itself…
And I would find stupid to have to create a new MR and to abandon the initial contribution…
I assume the problem would be the same on GitHub, right ?
Thanks for reading me, have a nice day.
Regards,
--
Vincent LORENZO
_______________________________________________
eclipse.org-committers mailing list
eclipse.org-committers@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/eclipse.org-committers
|
